Energy Minimization Methods in Computer Vision and Pattern Recognition

The book "Energy Minimization Methods in Computer Vision and Pattern Recognition" documents the contributions of the sixth International Conference on Energy Minimization Methods in Computer Vision and Pattern Recognition (EMMCVPR 2007), which took place from August 27 to 29, 2007, in Ezhou, Hubei, China. This conference addressed the realization that many challenges in computer vision and pattern recognition can be formulated through probabilistic inference or the optimization of energy functions. The topics covered include, among others, probabilistic grammars, image analysis, and the use of datasets with ground truth for evaluating algorithms as well as for training learning algorithms. The conference attracted 140 submissions, of which 22 were selected for oral presentation and 15 for poster presentation. The papers gathered in this book are divided into seven sections that deal with algorithms, applications, image processing, motion, shape, and three-dimensional processing.